BRITISH STANDARD BS 3580:1964

CONFIRMED
AUGUST 1985

Guide to design considerations on

The strength of screw

threads
BS 3580:1964

Co-operating organizations

The Mechanical Engineering Industry Standards Committee, under whose

supervision this British Standard was prepared, consists of representatives
from the following Government departments and scientific and industrial
organizations:

Admiralty* Gas Council

Air Ministry
Associated Offices’ Technical Committee
Association of Consulting Engineers
(Incorporated)
Association of Mining Electrical and
Mechanical Engineers
British Chemical Plant Manufacturers’
Association
British Compressed Air Society
British Electrical and Allied Manufacturers’
Association*
British Gear Manufacturers’ Association
British Internal Combustion Engine
Manufacturers’ Association
British Iron and Steel Federation
British Mechanical Engineering Federation*
British Railways Board*
Crown Agents for Oversea Governments and
Administrations
D.S.I.R. National Engineering Laboratory*
Electricity Council, the Generating Board and
the Area Boards in England and Wales
Engineering Equipment Users’ Association*
High Commission of India
Institute of Marine Engineers
Institute of Petroleum
Institution of Civil Engineers*
Institution of Gas Engineers
Institution of Heating and Ventilating
Engineers
Institution of Mechanical Engineers*
Institution of Mechanical Engineers
(Automobile Division)
Institution of Production Engineers*
Locomotive and Allied Manufacturers’
Association of Great Britain*
London Transport Board*
Machine Tool Trades Association
Ministry of Labour (H.M. Factory Inspectorate)
Ministry of Power
Ministry of Public Buildings and Works
Ministry of Transport
National Coal Board
National Physical Laboratory (D.S.I.R.)*
Radio Industry Council*
War Office*

The Government departments and scientific and industrial organizations

marked with an asterisk in the above list, together with the following were
directly represented on the Committee entrusted with the preparation of this
standard:

Agricultural Engineers’ Association Scientific Instrument Manufacturers

British Bolt, Nut, Screw & Rivet Federation
British Ship Research Association
Electronic Engineering Association
Ministry of Aviation
National Association of Marine Engine
Builders
Association
Screw Thread Tool Manufacturers Association
Society of British Aircraft Constructors
Society of Motor Manufacturers & Traders Ltd.
Telecommunication Engineering &
Manufacturing Association

This Guide, having been

approved by the Mechanical
Engineering Industry Standards
Committee and endorsed by the
Chairman of the Engineering
Divisional Council, was
published under the authority
of the General Council on
28 February 1964 Amendments issued since publication

© BSI 12-1999 Amd. No. Date Comments

The following BSI references

relate to the work on this
standard:
Committee references MEE/1,
MEE/1/2
Draft for comment A(MEE) 9770

ISBN 0 580 35254 4

 BS 3580:1964

Contents

Page
Co-operating organizations Inside front cover
Foreword ii
Introduction
1.1 Scope 1
1.2 Symbols 1
General
2.1 Introduction 1
2.2 Design principles 1
2.3 Form of bolt failure 2
2.4 Tensile strength of bolt related to stress area 2
2.5 Stripping strength 3
2.6 Fatigue strength 3
Materials
3.1 Tensile strength 3
3.2 Fatigue strength 4
Method of production
4.1 General 4
4.2 Cold rolling 4
4.3 Root-rolling 4
General form of threaded members and type of loading
5.1 Introduction 4
5.2 Stripping strength 5
5.3 Thread load distribution and fatigue strength 5
5.4 Impact strength 8
Diameter, pitch, D/p ratio, and length of engagement
6.1 Stripping strength 8
6.2 Tensile strength 9
6.3 Fatigue strength 9
Thread form
7.1 Stripping strength 10
7.2 Fatigue strength 10
Depth of engagement, degree of fit and truncation of threads
8.1 Stripping strength 11
8.2 Fatigue strength 11
Friction conditions
9.1 Introduction 12
9.2 Effect of torsional stress 12
9.3 Clamping load 12
9.4 Fatigue strength 13
Appendix A Formulae for thread shear areas and length of engagement 14
Appendix B Torque-tension relations 15
Appendix C Symbols 17
Appendix D References 18
Figure 1 — Forms of nut giving improved thread load distribution 6
Figure 2 — Theoretical load distribution along thread helix 7

© BSI 12-1999 i
BS 3580:1964

Foreword

This guide makes reference to the following British Standards:

BS 1083, Precision hexagon bolts, screws and nuts.
BS 1580, Unified screw threads — Parts 1 and 2: Diameters ! in and larger.
BS 1768, Unified precision hexagon bolts, screws and nuts (UNC and UNF
threads) — normal series.
BS 3139, High strength friction grip bolts for structural engineering —
Part 1: General grade bolts.
BS 3294, The use of high strength friction grip bolts in structural steelwork —
Part 1: General grade bolts.
This guide has been prepared under the authority of the Mechanical Engineering
Industry Standards Committee to serve as a guide to the design of load-bearing
screwed connections. Its aim is to make designers aware of the various factors
influencing the strength of such connections and to give some estimate of the
effect of these factors.
As the present degree of knowledge in the field is insufficient to warrant the
inclusion of precise formulae of general applicability, no attempt has been made
to present detailed calculations. Where values or formulae have been given, it
should be appreciated that these represent the best approximations which can be
given in the light of available knowledge. It is hoped that they will be refined and
extended in future revisions of this guide as a result of further investigation.
Permission to reproduce from the N.B.S. Handbook H 28 (1957) Part 1, text
relating to thread shear areas and lengths of engagement, has been kindly
granted by the National Bureau of Standards of America. This text is included
herein as Appendix A, to be used as an interim measure, to calculate the static
stripping strength.
Appendix B concerns torque-tension relations and the effect of thread friction on
permissible tightening stresses.
NOTE In place of the customary, but incorrect, use of the ton as a unit of force, the unit called
ton-force (abbreviation tonf) has been used in this guide.
A British Standard does not purport to include all the necessary provisions of a
contract. Users of British Standards are responsible for their correct application.
Compliance with a British Standard does not of itself confer immunity
from legal obligations.

Summary of pages
This document comprises a front cover, an inside front cover, pages i and ii,
pages 1 to 19 and a back cover.
This standard has been updated (see copyright date) and may have had
amendments incorporated. This will be indicated in the amendment table on the
inside front cover.

ii © BSI 12-1999
 BS 3580:1964

Introduction
1.1 Scope
This guide relates to applications of triangular (V-form) screw threads where strength considerations have
to be borne in mind. It is intended to draw the attention of designers to the principal strength factors to be
considered in following out details of a design, with some mention of measures which may be adopted to
improve strengths. The guide has been largely based on results obtained with steel components and while
many of the considerations will apply to other materials, care should be exercised in the interpretation of
certain data. For quantitative data, it will often be necessary to refer to an appropriate source of
information, such as a component or material manufacturer or a Government laboratory, though it is
emphasized that information on many aspects of the subject is still inadequate.
The guide also relates to behaviour at temperatures where creep and metallurgical changes are
unimportant and is concerned with screwed connections, i.e., with “bolt” and “nut” combinations; the
strength of a threaded bar, not assembled with a nut, is not considered. For the latter, reference should be
made to appropriate theoretical and experimental work on notched and threaded bars (1) (2)1).
NOTE “Nut” and “bolt” are used throughout in the general sense to mean internally and externally threaded members respectively,
except where it is obvious that ordinary nuts and bolts are meant.
The effect of the various strength factors are considered under the following headings:
Materials
Method of production
General form of threaded members and type of loading
Diameter, pitch, D/p ratio and length of engagement
Thread form
Depth of engagement, degree of fit and truncation of threads
Friction conditions.

1.2 Symbols
For ease of reference, symbols used throughout this guide are listed in Appendix C.

General
2.1 Introduction
To a given problem of thread design, there may be several solutions, between which it is not possible to
choose in the light of present knowledge. The preliminary choice of the general lines of a design must
therefore still be based, to some extent, on previous experience with similar problems.

2.2 Design principles

A threaded fastener will usually have to be designed to withstand axial loads, which may be static,
fluctuating, or impactive in nature. Supplementary bending and shear may be present; torsional loads will
arise mainly from thread friction on tightening and will be static in nature. The strength of a joint assembly
employing threaded fasteners will largely depend, particularly under fluctuating loads or shear loading
such as occur in structural steelwork2), on the overall design and provision of adequate pre-tensioning; the
latter will, of course, demand adequate static strength of the fasteners employed. Bearing this in mind, the
following is a discussion of the factors affecting the intrinsic strength of threaded connections, with only
brief reference to the effects of joint design.

1)
A list of references is given at the end of this guide.
2)
See BS 3139, “High strength friction grip bolts for structural engineering”. and BS 3294, “The use of high strength friction grip
bolts in structural steelwork”.

© BSI 12-1999 1
BS 3580:1964

It will be appreciated that the factors considered are often interrelated, e.g., the optimum tensile strength
of the nut material for a given bolt may depend not only on the tensile strength of the bolt material but also
on the pitch and diametral fit of the threads. At the same time, while it may be desirable to try to achieve
the optimum combination of materials and dimensions for a special application, it is necessary for ordinary
mass-produced bolts and nuts to use a restricted number of material combinations, each of which will have
to serve for a range of other variables, e.g. for various classes of fit. Thus, for instance, it may be
economically preferable to accept the strength of stock components, and to design accordingly rather than
to design for the higher strength available from the use of special components.

2.3 Form of bolt failure

It is desirable, where possible, to design a fastener so that failure under tensile load would occur by
breakage across the core of the bolt, rather than by thread stripping. The latter form of failure, which
begins by thread bending and ends by shearing of the internal and/or external threads, tends to be gradual
in nature, and progressive in cases of repeated assembly. Such damage is not always easy to detect,
particularly if the main damage is to internal threads; this introduces the possibility that serious
overtightening on assembly may remain undetected until evidenced by failure in service. Again, if failure
occurs by stripping, this indicates uneconomic use of the material of the bolt, the full core strength of which
is not developed.

2.4 Tensile strength of bolt related to stress area

Tensile strength of bolt. Where failure occurs across the core of the bolt, the tensile strength should be
computed as the product of the ultimate tensile stress of the material and the tensile stress area As: where
;
As = --- (Mean of effective and minor diameters)2
4
;
= ------ (Effective diameter + minor diameter)2.
16

The use of the stress area As has been found to give a reasonable approximation to the condition which
prevails at the point of fracture.
Tensile stress areas for Unified threads. The tensile stress areas for Unified threads, which are quoted in
BS 1580-13) are calculated by the above formula4) using basic effective and design minor diameter.
For ! in diameter threads, Class 1A, in the minimum metal condition, the stress area is less than the
quoted values by only about 8 per cent for UNC and 6.5 per cent for UNF and UNEF. This difference
decreases with increasing diameter and for 1" in threads is only about 3 per cent for UNC and 1 per cent
for UNF and UNEF. The corresponding differences for Class 2A threads are about 1 per cent less than
those for Class 1A. If such differences are considered to be important in a particular application, design
should be based on minimum metal dimensions for the class of thread employed.

3) BS 1580, “Unified screw threads”, Parts 1 and 2, “Diameters ! in and larger”.

4)
For Unified external threads,
17
H = 0.86603p, then E s = ( D – 3⁄4 H ) , K s =  D – ------ H and the formula reduces
12
2 2 2
---  E s – H
to A s = ; ----- = ;
--- ( E s – 0.28867p ) = 0.7854 ( D – 0.9382p ) . This formula correlates with test results for steels up
4 3 4
to 45 tonf/in2 tensile strength. For steels of greater tensile strength, the basic effective diameter should be replaced by the minimum
effective diameter for the class of thread in question.

2 © BSI 12-1999
 BS 3580:1964

2.5 Stripping strength

The stripping strength of a threaded combination is not easy to compute: formulae based on “shear areas”
are unrealistic, as they incorrectly assume that shear occurs in threads not previously deformed by
bending, and that the internally threaded member suffers no radial expansion; the latter assumption is
quite inaccurate for the lighter series nuts at failure loads; expansion will, of course, decrease with
increasing wall thickness. Again, if the stripping load is high enough to cause prior yield of the body of the
bolt, “necking” of the latter will reduce depth of engagement in a manner similar to that caused by nut
expansion, and plastic elongation of the bolt (increase of pitch), will necessitate severe deformation of
engaging nut threads, especially near the bearing face. Also, the stripping load of a given nut depends on
the hardness of the bolt; as this is increased, bending of the bolt threads at the nut failure load will be
reduced and shearing will take place nearer the root of the nut threads, which will increase the nut
stripping load. The foregoing argument will, of course, apply to the stripping strength of a bolt fitting into
a nut of harder material, a case which is sometimes unavoidable.
Despite the inadequacy of formulae based on the “shear area” of undistorted threads, this approach is at
present the only one generally applicable to the calculation of stripping strength, although the onset of
thread yield, which is due to bending, may be estimated from Sopwith’s analysis. (3) Until more
experimental data can be acquired, therefore, it is suggested that use be made of Appendix A to this guide,
which has been copied from pages 5 and 6 of the American “Screw-thread Standards for Federal Services”,
Handbook H.28 (1957), Part I, with some slight modifications and the addition of formulae for Whitworth
threads. Wherever possible, the values of critical length of engagement should be checked experimentally,
especially for the higher D/p ratios, where such experimental evidence as is available indicates that the
formulae give lengths of engagement which are too low.

2.6 Fatigue strength

Fatigue stresses quoted for screwed connections are usually nominal stresses computed on the core area of
;
the bolt, i.e. --- (minor diameter)2. Fatigue strength actually depends on the maximum true stress which is
4
much higher than the nominal, and on the stress distribution, which is non-uniform. (See 5.3). These are
difficult to estimate accurately and in any case bear no particular relation to the “stress area” for static
loading. It must be borne in mind that a fatigue strength quoted as a nominal core stress will not
necessarily apply to a threaded combination different from the one on which the determination was made,
due to differences in the actual stress distributions.
Some information on the fatigue strength of steel bolts, ! in to # in diameter, is contained in Ref. 21.

Materials
3.1 Tensile strength
Material. The nut material should, where possible, be somewhat softer than the bolt material. The ratio of
the tensile strength of the nut material to that of the bolt material, necessary to develop the full tensile
breaking load of the bolt, increases with diameter/pitch ratio. This is due to the lower stripping strength of
fine threads, as described in 6.1. For threads as fine as UNF this tensile strength ratio should not generally
be less than 0.85 when using a solid bolt, though a ratio of about 0.75 should suffice for threads near basic
dimensions (see also 6.2).

© BSI 12-1999 3
BS 3580:1964

Increased tensile strength. Increasing the tensile strength of the bolt material above about 60 tonf/in2 does
not usually result in much increase of fatigue strength in cases of severe stress concentration as obtained
in threaded connections (this does not necessarily apply to bolts with threads cold-rolled after heat
treatment where fatigue strength is largely dependent on the condition of the surface layer. See 4.2). Since
tight clamping can, however, ensure that the load in the bolt is increased by only a fraction of the load
applied externally to a joint assembly 5), high tensile strength bolts are often advantageous, particularly in
saving space and weight.
3.2 Fatigue strength
Increased fatigue strength. Use of a nut material with an elastic modulus lower than that of the bolt
improves the elastic thread load distribution. Thus, increases in fatigue strength of up to 40 per cent have
been reported for steel bolts assembled with cast iron or light alloy nuts, as compared with steel nuts. The
length of engagement in a relatively weak nut material must, of course, be great enough to provide
adequate static stripping strength.

Method of production
4.1 General
Effect of different machining methods. The various machining methods produce threads not varying greatly
in strength, provided that finish, and accuracy of thread form and pitch are comparable, and that faults,
such as grinding cracks or unfavourable residual stresses, are not introduced. The latter defects would have
an adverse effect on fatigue strength, while having little or no effect on static strength.

4.2 Cold rolling

Cold rolling can produce bolt threads with greater strength than machined threads. The static strength is
only increased by a few per cent but, in suitable materials, the fatigue strength can be greatly increased,
e.g., by 50 to 100 per cent for bolt steels. (9), (10) and (2) Chap. 8. Where applicable, this method is probably
the most effective one for increasing the intrinsic fatigue strength of a screwed connection. Some attention
must be paid to rolling conditions in order to obtain optimum results and further research is desirable on
the effect on fatigue strength of such factors as rolling time, rolling pressure and rate of penetration.
(See also 7.2). The increased strength, as compared with cut threads, is due to residual compressive stress
in the thread roots, together with work hardening and improvement of grain flow. Subsequent heat
treatment will reduce the fatigue strength again in the measure of the severity of this treatment,
e.g. complete re-hardening and tempering will reduce the fatigue strength to that of a thread cut after the
same heat treatment, whereas a low temperature stress relief may only cause a moderate reduction from
the optimum due to loss of some residual compressive stress, the favourable grain flow being retained.
Excessive decarburization will reduce the static stripping and fatigue strengths for any method of
production.

4.3 Root-rolling
Improvements similar to those for full-form rolling can be obtained from root-rolling of pre-cut
threads. (10) This method is useful for special threads and large diameters.
General form of threaded members and type of loading
5.1 Introduction
Form of fasteners. Apart from the overall design of the joint, the general form of the threaded members
themselves has some effect on their strength.

5)
For an external load P applied to the outer ends of the clamped members, the resultant load in the
 CB 
bolt is P B = Po + P  ------------------------- , where Po is the initial load in the bolt developed by tightening the nut and CB, CM are
 C M + C B
respectively the stiffnesses of the bolt-nut combination and of the clamped members (loads per unit deflection along the bolt
 C M + C B
axis). This applies for values of P where the load for joint separation, which is equal to Po  -------------------------  and is proportional to Po, is
 CM 
greater than P. Where the separating load is less than P, the whole of the external load is borne by the bolt.
For further discussion of this matter see refs. (4), (5), (6), (7), (8).

4 © BSI 12-1999
 BS 3580:1964

5.2 Stripping strength

Increased stripping strength. The static stripping strength of a connection can be increased by increasing
the external diameter or width across flats of the nut up to about 1.9 D or 1.8 D respectively for round and
hexagon nuts, though the increase in strength is small for dimensions greater than those of normal size
nuts (e.g., to BS 1083 or BS 1768)6). Stiffening the nut reduces its radial expansion under load; such
expansion causes loss of diametral fit, the effect of which is discussed under 8.1 [see also ref. (13)].

5.3 Thread load distribution and fatigue strength

Effect of bolt and nut diameters: The thread-load distribution is made only slightly more uniform by
increasing the nut diameter or hexagon above normal size and only slightly less uniform by cylindrical
boring out of the bolt up to about 0.6 D7). For a given load, the resultant stress at the thread root is due to
a combination of a, the thread bending stress; b, the tensile stress in the bolt core, concentrated by the notch
effect of the thread. On boring out the bolt, the maximum value of b is increased in proportion to the
reduction of core area, but that of a is only slightly increased due to slight worsening of the thread-load
distribution. On boring out, therefore, the increase in resultant maximum stress is not proportionately as
great as the increase in nominal stress, based on core area; i.e. in a hollow bolt, there is a better utilization
of material than in a solid bolt. Thus a hollow bolt, bored out to a diameter not greater than 0.6 D, and
having a weight equal to x per cent of that of a comparable solid bolt of the same external diameter, will
have a limiting fatigue load greater than x per cent of that of the solid bolt.
As a special measure, however, conical boring out of the bolt materially improves the elastic thread load
distribution, as does conical tapering of the outside of the nut, especially when this is combined with the
“tension” or “overhung” nut design (2), (3), (11), which has the effect of putting the threaded portions of both
bolt and nut under tension. Examples are shown in Figure 1 a and Figure 1 b.
Forms of thread load distribution. A tension-tension, or turnbuckle type of connection has, in any case, a
better thread load distribution than an ordinary nut and bolt connection: it has, in fact, the best load
distribution of all connections, classified according to loading of members. The order of merit in this respect
is, beginning with the most favourable:
A) nut in tension, bolt in tension,
B) nut in shear (i.e. held at the sides), bolt in tension,
C) nut in compression, bolt in compression,
D) nut in compression, bolt in tension,
E) nut in shear, bolt in compression,
F) nut in tension, bolt in compression.

6)
BS 1083, “Precision hexagon bolts, screws and nuts”.
BS 1768, “Unified precision hexagon bolts, screws and nuts (UNC and UNF threads) — normal series”.
7) Boring out the whole shank reduces the axial stiffness of the bolt which may improve the behaviour of the complete joint
(see footnote to 3.1). As compared with reduction of the external shank diameter (waisting), boring out preserves greater
torsional and bending moments of resistance for the same cross-sectional area; reduced bending stiffness may, however, be
desirable to prevent excessive bending loads being transferred to the threaded portion.

© BSI 12-1999 5
BS 3580:1964

NOTE These diagrams do not claim to show optimum proportions.

Figure 1 — Forms of nut giving improved thread load distribution
For a given case in the UNC or UNF series, the maximum elastic thread load per unit length of helix is
over 80 per cent greater for case F) than for case A); for the common case D) of an ordinary nut and bolt, it
is 40 to 45 per cent greater than for case A), [from work by R. M. Gray (17), Purdue University, U.S.A., based
on Sopwith’s analysis (3)]. Similar relations would hold for Whitworth or Metric threads. Typical thread
load distributions are given in Figure 2 for 1-12 UNF threads, for a coefficient of friction of 0.3. For coarser
threads or lower coefficients of friction, the load distributions are somewhat more uniform, but the curves
for the various conditions of loading A) to F) bear a similar relation to each other.
Improvement of thread load distribution. The thread load distribution and hence fatigue strength may also
be improved by tapering the effective diameter of the thread of the nut and/or bolt, but this again is a
special measure, not practicable for smaller diameters. (3) (12) The degree of taper required is a few
thousandths of an inch on diameter per inch of length of engagement, depending on the load at which
optimum thread load distribution is required.

6 © BSI 12-1999
 BS 3580:1964

(From R. M. Gray: “Preliminary investigation of load distribution in threaded connections”, 1954, Division of Engineering Science
Laboratory, Purdue Research Foundation, Lafayette, Indiana U.S.A.)
Figure 2 — Theoretical load distribution along thread helix
The general idea behind such methods is to increase the deflection per unit thread load of those threads
which are normally most heavily loaded, compared with that of the other threads. The improvement in
thread load distribution so obtained outweighs the adverse effects of increasing the bending stress per unit
thread load on the threads which were originally most heavily loaded. It is possible, by special machining,
to reduce the stiffness of the latter threads by progressive removal of metal from the thread roots or flanks,
maximum removal occurring at the end of the helix which is normally most heavily loaded. In doing this,
care should be taken not to introduce notches which are more severe than the original thread grooves.

© BSI 12-1999 7
BS 3580:1964

Measures adopted to improve the thread load distribution in an ordinary nut and bolt type of combination
should aim at producing a thread loading near the free end of the nut which is somewhat greater than that
near the bearing end, since the tensile stress in the body of the bolt decreases towards the free end of the
nut. The resultant stress at the thread root is due to a combination of the thread bending stress with the
tensile stress due to the load transmitted across each section of the bolt. The latter stress is concentrated
in the region of the thread roots by the geometrical notch effect but, of course, falls off towards the free end
of the nut. In a combination with an inverted thread load distribution as described above, the resultant
distribution of maximum stress in the threads could be made approximately uniform.
To obviate fatigue failures at the thread run-out, flat-bottomed threads should be avoided at this situation,
and the run-out should not occur too sharply. It may be eliminated by a smoothly radiused relieving groove.
After assembly, the run-out should not be near the bearing face of the nut; this is in order to avoid
superimposing the stress concentration due to the run-out on that due to the concentrated loading on the
first threads in engagement. Extensive waisting of the bolt shank is not superior to a short well-radiused
groove as regards the intrinsic fatigue strength of the bolt, but may considerably increase the fatigue
strength of a complete joint by decreasing the ratio of the stiffness of the bolt to that of the clamped parts,
thereby reducing, in the presence of adequate pre-tightening, the fraction of any external load, by which
the bolt load is increased8). Flexible bolt heads with recesses or slots are commercially available; these are
especially useful in reducing the stiffness of short bolts or screws.

5.4 Impact strength

As with fatigue strength, the impact strength of a joint depends principally on the overall design, which
should aim at absorbing as much energy as possible in unnotched parts.
A bolt itself may be waisted9) to absorb energy by stretching the shank rather than the threaded portion.
The greater the length of shank between the bolt head and the nut the greater the impact strength, and
this applies also to bolts threaded up to the head. For waisted bolts, the shank area to give maximum
impact strength is about equal to the tensile stress area of the threads, or about 10 per cent more than this
if the threads have been cold rolled and not subsequently heat treated.
It is important to avoid notches and sudden large changes of section in the various members of the joint.

Diameter, pitch, D/p ratio and length of engagement

6.1 Stripping strength
The stripping strength of a screwed connection of given diameter and length of engagement decreases with
pitch (i.e., with increasing fineness) but is not simply proportional to p. The relation between stripping
strength and D/p varies with the ratio of the tensile strengths of the nut and bolt materials. No general
relationship can be given on the basis of existing data.
The critical length of engagement Lcrit, required to develop the full tensile strength of a given bolt, is
inversely proportional to the stripping strength per unit length of engagement. Total stripping strength is
proportional10) to Le and will determine the strength of the connection if Le is less than Lcrit. For a thread
of given diameter, Lcrit increases as the pitch decreases, due to decrease of stripping strength per unit
L
crit
length of engagement and to increase of bolt core area. For a given D/p, the ratio ------------- decreases as the
D
diameter increases and, as with stripping strength, Lcrit is not simply a function of D/p.
If Lcrit is known in a given case, it follows that the stripping strength for lengths of engagement less than
Lcrit can be found, as the stripping strength for Lcrit is equal to the tensile strength of the bolt and stripping
strength is proportional to L10).
For such applications as thin-walled pipe joints, radial expansion of the nut and contraction of the bolt will
reduce the stripping strength below that of similarly threaded thick-walled members, though the necessary
length of engagement will be limited by the tensile strength of the (hollow) bolt.

8)
See footnote to 3.1.
9)
See footnote to 5.3.
10) Thisis true for values of L up to a certain limit which depends on the material combination. For steel nuts and bolts this
limiting value of L is probably not less than 1.2 D. [See ref (13)].

8 © BSI 12-1999
 BS 3580:1964

It should be noted that UNF threads are finer and more liable to fail by stripping for a given length of
engagement, etc., than BSF threads. This is in no way a deficiency of Unified threads as such, but simply
a result of the difference in pitch of these particular series. The static failure load, due to a single loading,
for a UNF bolt with an ordinary nut, e.g., to BS 1768,11) (height = 0.8 D to 0.9 D), is at least as high as that
of a corresponding BSF combination to BS 108312) it is the mode of failure which makes the UNF thread
less suitable for some applications than coarser threads (see 2.3). Where thread stripping is a problem,
UNC would be more suitable than UNF in some cases formerly employing BSF. UNF is not recommended
for ordinary bolts and nuts in sizes over 1 in, owing to the constant pitch and rapidly increasing D/p ratio.

6.2 Tensile strength

It is generally possible to develop the full tensile strength of a bolt by using a sufficient length of
engagement. Increase of the latter above about 0.6 D does not, however, result in any material decrease in
the bending stresses in the most heavily loaded threads near the bearing face of the nut, so that these
threads in the nut and/or bolt undergo plastic bending at approximately the same load for any length of
engagement above 0.6 D. For cases of repeated assembly this is undesirable and it is better in such cases
to achieve adequate stripping strength by using a nut material of sufficient tensile strength rather than to
increase the length of engagement above the normal value (e.g., approximately 0.8 to 0.9 D for steel nuts).
It is, however, inexpedient to increase the tensile strength of the nut material above, say 0.8513) times that
of the bolt material, as this will merely transfer stripping failure to the bolt threads without increase of
failure load.
Trouble may be experienced with UNF threads if they are tightened or otherwise loaded to yield,
particularly in cases of repeated assembly. Plastic thread bending begins in a UNF bolt at a load about
equal to or somewhat below that required to cause tensile yield of the body of the bolt (depending on the
class of fit, etc.). While this thread bending may be acceptable, if not too severe, in applications requiring
only a single tightening, it is undesirable in cases of repeated assembly, as it leads to progressive thread
failure. Account should be taken of this in designing and pre-tensioning; the maximum working stress
should be somewhat below the yield stress, as calculated on the tensile stress area of the bolt and after
making allowance for the effect of frictional torque as described in 9 and Appendix B.

6.3 Fatigue strength

For values of D/p up to about 18 and diameters up to about 3 in, the evidence shows that, for machined
threads, the effect of pitch on the fatigue strength of a bolt of given diameter is small. (See also 7.2 in
relation to rolled threads). The results of one investigator (14) on bolts of 90 mm (3" in) diameter with
machined threads, suggest that there is an optimum pitch (in this case corresponding to 4" to 5 t.p.i.) for
larger diameters.
Theoretical estimation of the effect of pitch on fatigue strength is difficult, owing to the lack of coincidence
of the maximum stress due to thread bending, which occurs near the junction of thread flank and root
radius, and the maximum stress due to the tensile load in the bolt and the thread notch effect, which occurs
at the bottom of the bolt thread root. The latter decreases with decrease of pitch, since the nominal stress
decreases with increase of core area and since, for a given form of thread, the notch stress concentration
factor does not vary greatly with pitch (as this is much less than D). The maximum stress due to thread
bending, resulting from transference of load from nut to bolt, increases, however, with decrease of pitch,
owing to the increase in concentration of thread load along the helix. As far as is possible to estimate, these
two opposing factors approximately cancel out, at least for a diameter/pitch range covering the series as
stated above. This conclusion is confirmed by the limited amount of experimental evidence available.
For very fine threads, fatigue failure of the individual threads may occur, resulting in breaking off of pieces
of thread at lower loads than would be necessary for cracks to develop across the bolt core; this has been
known to occur, for instance, for a diameter/pitch ratio of 20 at about + in diameter, i.e., slightly finer than
UNEF at this size.

11)
BS 1768, “Unified precision hexagon bolts, screws and nuts (UNC and UNF threads) — normal series”.
12)
BS 1083, “Precision hexagon bolts, screws and nuts (BSW and BSF threads)”.
13)
Existing data are insufficient to give accurate estimates covering all cases. Stresses in the nut are lower than in the bolt.

© BSI 12-1999 9
BS 3580:1964

Thread load distribution may be improved by using a slightly greater pitch for the nut than for the bolt
thread (3). This is not normally practicable, except for large diameters, but might be achieved by machining
under load.
Serious pitch errors introduce high thread load concentrations and reduce fatigue strength, but small
errors of pitch such as may be encountered in commercially produced threads do not appear to have much
effect on fatigue strength (10).
There is a moderate size effect, e.g., bolts of 2" in diameter in a 35 tonf/in2 steel have a fatigue limit of the
order of 10 per cent less than geometrically similar bolts of # in diameter. Further reduction of the fatigue
limit with increasing size is not likely to exceed another 10 per cent for low tensile steels. There appear to
be no similar data for high tensile steels; by analogy with other forms of stress concentration, the full
reduction in fatigue strength due to “size effect” would be expected to be achieved at lower diameters,
perhaps in the region of 2 in. The above does not take into account any metallurgical differences associated
with size.
As elastic conditions are critical in fatigue, the intrinsic fatigue strength of a screwed connection is not
much increased by increasing the length of engagement beyond 0.6 D (this may be necessary, however, for
reasons of static strength and pre-tensioning), as the maximum thread stresses are not appreciably
affected thereby.

Thread form
7.1 Stripping strength
Thread form, over the range of triangular threads in general use (Unified, B.A., Whitworth, American
National, Metric, etc.) has no great effect on the static stripping strength of a connection, and a negligible
effect on the bolt tensile strength, except insofar as the stress area may differ for a given size.
Errors of thread form may reduce stripping strength if they involve loss of metal at the thread flanks or
roots. The effect of truncation is dealt with separately in 8.

7.2 Fatigue strength

For the same root radius and depth of engagement, the effect of thread angle on fatigue strength is small (9).
The root radius of the bolt should be of smooth form and blend smoothly with the thread flanks. Sharp
corners in the root region, where the maximum stresses occur, cause high local stresses and reduce fatigue
strength. The preferred root form is a single circular arc, as in the basic Whitworth form or in the design
form of the Unified external thread in BS 1580. The root radius of the nut is not normally important, but
proper radiusing should be provided where stresses are critical, as in thin-walled components subjected to
high fluctuating stress, or in “notch-sensitive” materials.
For bolt threads produced by cutting or grinding, or subjected to complete re-hardening and tempering
after rolling, the effect of varying the root radius from basic to half basic, either for Whitworth or for Unified
threads, is to reduce the fatigue strength by 15 to 20 per cent, for both low and high tensile steels (9).

10 © BSI 12-1999
 BS 3580:1964

Available evidence (9) indicates that a similar variation of root radius has little effect on finish cold-rolled
(i.e., not subsequently heat treated) threads of low or medium tensile steel (up to 45 tonf/in2). As mentioned
in 4.2, however, the effect of rolling conditions on fatigue strength, as distinct from correct size and thread
form, are not yet fully understood. The fatigue strength of finish cold-rolled high tensile steel bolts
increases rapidly with increase of root radius. This means that, for finish cold-rolled threads, larger pitches
may give greater fatigue strength, since the effect of increased root radius will probably outweigh any effect
of pitch as such14). To obtain maximum benefit from cold-rolling of high tensile steels, the root radius
should be kept as close to the basic value as possible. Where it is unnecessary to use nuts of minor diameter
close to the standard basic value, the root radius of the bolt thread may even be increased with advantage
to a value greater than basic. The minimum minor diameter of the nut to be used with such a bolt should
be somewhat greater than the diameter corresponding to the transition from flank to root radius of the bolt
thread, in order to avoid superimposing contact stresses on the maximum thread bending stresses. Such
increase of root radius above basic is probably only worthwhile in finished cold-rolled bolts and on very high
tensile bolts (70 tonf/in2 and above), intended for applications involving fatigue loads.
Errors of thread form, like those of pitch, do not generally have much effect on fatigue strength. There is
some evidence that any such effect is less for threads having substantial clearances on effective and crest
diameters than for threads near to basic dimensions [ref. (2), Chapters 4 and 5]; it must, however, be
remembered that close fit threads would be less likely to have thread form errors and would have greater
static stripping strength than free fit threads.

Depth of engagement, degree of fit and truncation of threads

8.1 Stripping strength
The static stripping strength of a connection depends markedly on the effective diameter clearance. Thread
crest truncation has not such a marked effect in practice on stripping strength despite the implications
of Appendix A.
For a nut and bolt, both of low tensile steel, if ¹eE and ¹eT are respectively half the effective diameter
clearance and half the truncation on the thread crest diameter, both expressed as a percentage of basic
depth of engagement, then the corresponding reduction in stripping strength, expressed as a percentage of
the static stripping strength of basic size threads, have been found to be respectively about 0.65 ¹eE
and 0.2¹eT [based on ref. (16)]. These figures apply at least to the minimum depths of engagement permitted
by BS 158015). They should be applicable to other cases where the nut and bolt materials are similar to each
other.
In order to ease production, BS 158015) permits increase of nut minor diameter tolerances where the length
of engagement exceeds 1" D, “provided that the actual length of engagement is substantial”. In
interpreting this last phrase, account should be taken of the fact that, for such long lengths of engagement,
the stripping strength of a combination is less than would be given by a linear extrapolation from values of
stripping strength valid for normal lengths of engagement. (See footnote to 6.1.)
Even as a special measure maximum nut minor diameter and minimum bolt major diameter limits should
not be exceeded on threads as fine as UNF, particularly for repeated assembly applications, as there is
already a tendency for failure to occur by thread bending and stripping in such fine threads.
It should be borne in mind that, where depth of engagement may be lost by such processes as wear and corrosion, the percentage
loss of depth of engagement and hence of stripping strength will be less for coarse than for fine threads.

8.2 Fatigue strength

In contrast to static stripping strength, fatigue strength is not much affected by degree of fit (see also 7.2).
It should be noted, however, that a commercial “free fit” thread may have other errors which do affect
fatigue strength, e.g., bad root form.
Reduction of depth of engagement by thread crest truncation, within the limits permitted by current
standards, has no adverse effect on fatigue strength. [Ref. (2) Chap. 4], (9).

14) Compare, for instance, the result on rolled threads of H. de Leiris (15) with his results on machined threads (14).
His 64 mm (2" in) machined thread bolts with 5 mm and 6 mm pitch had the same fatigue strength, whereas for rolled threads
of the same size, the 6 mm pitch bolts were 13 per cent stronger than the 5 mm pitch bolts.
15)
BS 1580, “Unified screw threads”, Parts 1 and 2, “Diameters ! in and larger”.

© BSI 12-1999 11
BS 3580:1964

A 20 per cent increase in fatigue strength has been reported due to conical boring out of the nut threads at
an included angle of 20°, to a maximum equal to the major diameter at the bearing face. Such heavy
truncation can be used to increase fatigue strength, even without taper boring, owing to improved thread
load distribution resulting from local thread indentation under the high contact pressures involved. For
this purpose, it is necessary to reduce the depth of engagement to about 25 per cent of basic, and the nut
should be softer than the bolt which should have threads with rounded crests. Such heavy truncation tends,
however, to cause seizure on tightening and is not suitable for cases of repeated assembly or where
torque-tension relations are used. Stripping strength is also somewhat reduced as noted above.
Friction conditions
9.1 Introduction
Thread friction has an effect on static strength insofar as it causes the superimposition of torsional stress
on the tensile stress in the bolt.
In discussing the ultimate tensile and yield or proof loads of a fastener employed in an axially loaded bolted
joint, a distinction must be drawn between the case of tightening the fastener itself to yield or failure on
assembly, and the case of load applied to the clamped parts, after initial tightening of the fastener within
its elastic range.

9.2 Effect of torsional stress

In the first case, owing to the torsional stress induced by thread friction and, to a smaller extent, by the
circumferential component of the thrust on the thread helix, the axial tensile component of the stress in
the bolt, required to produce yield or failure, will be less than that obtained in a pure tensile loading. For
combined stresses, the criterion of yield is the effective or comparative stress Bv, which is a function of both
tensile stress B and shear stress E such that .
In the second case, the effect of such torsional stress on bolt yield will depend on the degree of tightening,
affecting the torque developed in the bolt, and on the possibility of torque release due to rotational slip of
the bolt head or nut, or to rotation of the clamped parts relative to each other. The tensile strength of the
joint will, in this case, be the normal tensile strength of the fastener; joint separation, due to plastic
deformation of the fastener, will occur before final failure and the stress will be redistributed even without
rotation of the joint components.

9.3 Clamping load

For a joint tightened to yield of the fastener, the clamping load developed will be, for normal, lubricated
threads where the coefficient of friction is, say, from 0.2 to 0.15, about 70 to 80 per cent of the normal
proof16) or yield load of the fastener as obtained under pure tension; in other words, only 70 to 80 per cent
of the tensile proof load can be regarded as available for clamping purposes. If the coefficient of friction is
reduced to 0.1, this figure is increased to 90 per cent; for absolutely dry, unlubricated threads, without
special surface treatment, it may be reduced to 50 per cent, and will vary widely due to slight differences
in the surface conditions. The increased thread friction in the latter case will require higher values of
tightening torque. Tightening torques are only a useful indication of the clamping load when the frictional
conditions (including those at the nut bearing face) are not unduly variable and correspond to those for
which the torque-tension relation was established. Some surface treatments, e.g., cadmium plating, reduce
the coefficient of friction. Lowest and most consistent values of the coefficient of friction are obtained for a
thin plating applied either to the nut or to the bolt rather than to both; the value and consistency of the
coefficient of friction for such a combination in the dry condition are similar to those for lubricated threads
without surface treatment. Special lubricants, such as molybdenum disulphide grease, are more effective
than oil in reducing friction and will give higher clamping loads for a given tightening torque; care should
be taken in using them to adjust the assembly torque correspondingly. For a fuller discussion of tightening
relations, see Appendix B.

16)
For a threaded member, load for 0.5 per cent permanent strain.

12 © BSI 12-1999
 BS 3580:1964

Control of clamping load without torque measurement may be achieved, according to the application, by
measuring bolt elongation, by use of commercially available, load-indicating bolts or washers, (18) (19), or by
turning the nut a given amount after initial pre-seating (18) (20).

9.4 Fatigue strength

As has already been mentioned in 5.3, reduction of thread friction improves the thread load distribution (3)
and may be expected to have a beneficial effect on fatigue strength; in the course of prolonged fatigue
loading, however, unless the lubricant or surface treatment employed is able to withstand the fretting
action between the threads, this benefit will probably disappear.

© BSI 12-1999 13
BS 3580:1964

Appendix A Formulae for thread shear areas and length of engagement

For interim use in the calculation of static stripping strength. Taken from the National Bureau of
Standards (U.S.A.) Handbook H.28 (1957), Part I, “Screw Thread Standards for Federal Services”, pages 5
to 6, with slight modifications and the addition of formulae covering Whitworth threads.
The thread shear area of the external thread is the effective area at a diameter equal to the maximum
minor diameter of the internal thread. The thread shear area of the internal thread is the effective area at
a diameter equal to the minimum major diameter of the external thread. The formula for shear area of the
external thread at a diameter equal to the maximum minor diameter of the internal thread (ASs) is as
follows:
For Unified threads:
1
AS s = 3.1416 nL e K n, max. ------- + 0.577 ( E s, min. – K n, max.)
2n
For Whitworth threads:
1
AS s = 3.1416 nL e K n,max. ------- + 0.521 ( E s, min. – K n, max.)
2n
The formula for shear area of the internal thread at a diameter equal to the minimum major diameter of
the external thread (ASn) is as follows:
For Unified threads:
1
AS n = 3.1416 nL e D s, min. ------- + 0.577 ( D s, min – E n,max. )
2n
For Whitworth threads:
1
AS n = 3.1416 nL e D s, min. ------- + 0.521 ( D s , min. – E n,max. )
2n
As materials bearing the same name vary greatly in ultimate strength and in other essential
characteristics, the formulae given below are included in order that a safe length of external thread mating
with internal threads may be calculated. It is desirable that the length of internal thread and the
dimensions of this thread, particularly its minor diameter, be such that, taking into account a possible
difference in strength of material of the internal and external threads, the threaded portion of the external
thread will break before either the external or internal threads strip. For this reason, the shearing strength
of the material should be taken as one half the tensile strength, which gives a small factor of safety.
The length of engagement of a threaded unit, that will develop maximum strength of an assembled
threaded unit with external and internal threads manufactured of materials of equal tensile strength, is
computed from the following formula:
For Unified threads:
2 × stress area
L e = ------------------------------------------------------------------------------------------------------------------------------------------
1
3.1416 nK n, max. ------- + 0.577 ( E s, min. – K n, max. )
2n
For Whitworth threads:
2 × stress area
L e = ------------------------------------------------------------------------------------------------------------------------------------------
1
3.1416 nK n, max. ------- + 0.521 ( E s, min. – K n, max. )
2n
This formula has the factor “1/2” for relation of shearing strength to tensile strength incorporated therein.
The formula, while given for steel external and internal threads, may be used for brass external and
internal threads and provides an additional safety factor.
Where the external and internal threads are manufactured of materials of different tensile strengths, the
factor J for the relative strength in shear of external threads with respect to internal threads must be
considered. The factor J is computed from the following formula:
AS s × tensile strength of external thread material
J = -----------------------------------------------------------------------------------------------------------------------------------------
-
AS n × tensile strength of internal thread material

14 © BSI 12-1999
 BS 3580:1964

The length of engagement of a threaded unit adjusted to obtain proper relation of strength to cause
breakage of the bolt before threads will shear is Q and is computed from the following formulae:
If J is less than 1, Q = Le
If J is greater than 1, Q = J× Le

Appendix B Torque — tension relations

This appendix concerns torque-tension relations and the effect of thread friction on permissible tightening
stresses.
To a good theoretical approximation, the formula for the torque required to tighten a bolt and nut or a
machine screw, so as to develop a clamping load Po, is as follows:

. . (1)

Es
The torque is divided into three terms: P o ------
2
tan 2 represents that absorbed in driving the mating thread
helices over each other against the action of the axial load Po, to which they are inclined,
Es 41 do + d i
whilst P o ------ -------------- and P o -----------------
2 cos !
- 42 represent that absorbed in overcoming friction, in the first case
4
between the threads and in the second case at the bearing face under the nut or screw head.
In the ordinary nut and bolt case, a typical distribution of torque between the three terms would be, in the
above order, 10 per cent, 40 per cent and 50 per cent of the total. Friction conditions are therefore of
predominating importance, and unless they are known accurately, together with the nut or screw head
bearing area, there is no point in using an expression more complicated than the simple formula:
T = 1
--- P o D . . . . . . . . (2)
5

For threads which are lubricated with at least a thin film of ordinary oil or grease, formula (2) is generally
accurate to about ± 20 per cent. If this is not sufficient, it is recommended that torque-tension relations be
established experimentally for the conditions of the particular application. With the accumulation of
sufficient data and good production control, it may be possible to assign values to the terms of formula (1)
and set confidence limits for the torque values obtained from it.
The use of smooth, hardened washers will improve consistency if the nut or screw head abutment surfaces
are rough and give rise to variable friction. The use of a lock-washer with a serrated or otherwise uneven
face underneath the nut or screw head may make the torque-tension relation very variable and
experimental checks would be necessary. Further, bedding down or indentation under load of bearing faces
which are uneven due to use of serrated washers, to rough surface finish, or to lack of squareness, may
cause loss of tightness in service even though unscrewing may be prevented; in such cases tightness checks
using a torque spanner or other suitable means should be carried out until maintenance of tightness has
been confidently established. In general, the best method of maintaining the tightness of a stiff joint is by
adequate pre-tightening and provision of good bearing surfaces, without use of special nuts or washers. For
flexible joints, however, and cases where a loose nut or screw could cause serious damage, the use of special
locking devices is sometimes advisable. It should be borne in mind, however, that loss of tightness without
unscrewing could, under oscillating load conditions, cause the bolt to be subjected to fatigue loads great
enough to fracture it, (see footnote to 3.1) with the very result that the locking device was designed to
prevent. Locking devices without uneven bearing surfaces are, of course, available.

© BSI 12-1999 15
BS 3580:1964

In estimating safe tightening torque, account must be taken of the torsional stress in the bolt due to thread
friction and to the circumferential component of the thrust on the thread helix. The commonly used
criterion for yield under combined stress is the comparative stress:

Using this criterion, the axial stress in the bolt is:

. . . . . . . . . (3)

Only the first two torque terms in equation (1) contribute to the twist in the bolt, hence

P
Also B = -----o- , whence
As

Hence, substituting for --E- in equation (3), we have

B

If we let the permissible value of Bv be S times its value at yield of the bolt, i.e., the yield stress17) By of the
bolt material, then the permissible value of the axial stress is:

W may be taken as approximately equal to , using the same functional

diameter " (Ks + Es) as for calculating As, which is equal to ,

hence

17) If
By is determined on a plain unthreaded specimen, it is either the yield or 0.2 per cent proof stress of the material. If it is
determined on a threaded specimen, it is either the yield load or that required to cause 0.5 per cent permanent strain, divided by
the stress area As.

16 © BSI 12-1999
 BS 3580:1964

For bolts with a plain shank, which may be the weakest portion under static loads,
As = ; ;
--- ds2 and W = ------ds3 ,
4 16

whence

For machine construction, S may be taken as 0.918) for coarse threads such as UNC, BSW and BSF up
to 1 inch; for fine threads such as UNF or BSPF it would be safer to take S = 0.718) (see 6.2). This allows
for some additional load coming on the bolt due to service loads applied to the joint. If the clamped members
are relatively flexible, a smaller value of S may be necessary (see footnote to 3.1). In structural steelwork,
it has been found permissible to load the coarse-thread bolts employed up to yield. In such applications,
maximum clamping force is important in preventing slip between plates and girders, etc., and service
loading is likely to be mainly in shear.
A fuller discussion of the problem of tightening relations, together with experimental results, will be found
(in German) in the work of R. Kellermann and H.-Ch. Klein, surveyed by Klein with bibliography in
Technische Rundschau, 52 (47) 1960 Nov. 11th, Zweites Blatt pp. 9–13 and Drittes Blatt pp. 17–21.

Appendix C Symbols
As Tensile stress area
ASs Shear area of external thread at a diameter equal to the maximum minor diameter
of the internal thread
ASn Shear area of internal thread at a diameter equal to the minimum diameter of the
external thread

CB Stiffness of the bolt-nut combination 
CM Stiffness of the clamped members  axial loads per unit deflection

D 
Basic major diameter (i.e. nominal diameter, except for BSPF threads)
Ds, min. Minimum major diameter of external thread
do Diameter of outer edge of bearing area under the nut or machine screw head
di Diameter of inner edge of bearing area under the nut or machine screw head
ds Diameter of plain shank of bolt or machine screw
Es Basic effective (pitch) diameter of external thread
Es, min. Minimum effective (pitch) diameter of external thread
En, max. Maximum effective (pitch) diameter of internal thread
H Height of fundamental triangle
Ks Design minor diameter of external thread
Kn, max. Maximum minor diameter of internal thread
Lcrit Critical length of engagement
Le Length of engagement
n Number of threads per inch
P External load applied to clamped member
PB Resultant load in the bolt
Po Initial tensile load in the bolt or machine screw developed by tightening
p Pitch
S Safety coefficient
T Tightening torque

18)
These values should not be regarded as overriding accepted practice in individual industries.

© BSI 12-1999 17
BS 3580:1964

W ; 3
Polar moment of resistance of thread, taken as approximately equal to ---------
- ( Ks + Es )
128
! Flank angle of thread i.e., half the included angle
¹eE Half the effective diameter clearance
¹eT Half the truncation on the thread crest diameter
p - for single start thread
Thread lead angle: Tan 2 = -------
2
;E
l
Tan 2 = -------- for multi-start thread
;E
where l = lead
41 Coefficient of friction between mating threads
42 Coefficient of friction at bearing face of nut or machine screw head
B Axial tensile stress in bolt or machine screw
Bv Effective or comparative stress (used as the criterion of yield in the bolt or machine screw)
By Yield or proof stress of the bolt or machine screw material
E Shear stress (torsional stress in the bolt or machine screw)

Appendix D References
(1) Neuber, H., Theory of notch stresses; Edwards, Michigan, U.S.A., 1946. Second edition 1958, U.S. Atomic
Energy Commission (English translations from the German, Springer Berlin, 1937 and 1958).
(2) Jakushev, A.I., Effect of manufacturing technology and basic thread parameters on the strength of
threaded connections, (English translation from the Russian), Pergamon Press 1964.
(3) Sopwith, D.G., “The distribution of load in screw threads,” Proc. Inst. Mech. Engineers,
London 159, 1948, 373–383.
(4) Erker, A., “Design of screw fastenings subject to repeated stresses,” Proc. Internat. Conf. on Fatigue of
Metals, Inst. Mech. Engineers, London, 1956, 290–300.
(5) Boomsma, M., “Loosening and fatigue strength of bolted joints,” Engineer 200 (5195) 1955
Aug. 26th, 284–286. See also 203 (5284) 1957, May 3rd, 674–675.
(6) Gill, P. J., “Notes on the load carrying characteristics of pre-tensioned bolts — tensioned joints,” Inst. of
Struct. Engineers, Jubilee Symposium on high strength bolts, 1959, June 10th, 62–64.
(7) Field, J. E., “The preloaded bolted joint under applied axial load,” Engineering Designer, 1961, Aug. 3–6.
(8) Field, J. E., “Stiffness of plates in a bolted joint,” Engineering Designer, 1962, Aug. 18.
(9) Sopwith, D. G. & Field, J. E., “Unification of screw thread practice,” Engineer, 203 (5287) 1957,
May 24th, 793–795.
(10) Cook, R. & McClimont, W., “The influence of screw forming methods on the fatigue strength of large
bolts,” Trans. Inst. Marine Engineers, 73 (12) 1961, Dec., 417–429.
(11) Brown, A. F. C. & Hickson, V. M., “A photoelastic study of stresses in screw threads,” Proc. Inst. Mech.
Engineers (London) 1B (12) 1952-53, 605–612.
(12) Stoeckley, E. E. & Macke, J., “Effect of taper on screw thread load distribution,” Trans. Am.Soc.Mech.
Engineers, 74 (1) 1952, Jan., 103–112.
(13) Field, J. E. & Waters, D. M., “Effect of nut height on static stripping strength of soft aluminium nuts,”
NEL Rept. No. 23, Feb. 1962, National Engineering Laboratory, East Kilbride, Glasgow.
(14) De Leiris, H., “Etude expérimentale de l’influence du pas sur la résistance à la traction cyclique des
assemblages boulonnés de gros diamètre,” (Experimental study of the effect of pitch on the fatigue strength
in pulsating tension of large diameter bolt assemblies); Bull. Ass. Tech. Maritime et
Aéronautique, 57, 1959, 429–449.
(15) De Leiris, H., Filetages: Essais de traction cyclique sur filetages de 64 mm de diamètre obtenus par
roulage (Screw threads: Pulsating tension fatigue tests on 64 mm rolled threads), Report E O3-RE 3 of
AFNOR (Association Française de Normalisation), Jan. 1962.

18 © BSI 12-1999
 BS 3580:1964

(16) Smith, C. W. & Low, A. C., “Effect of fit and truncation on the strength of Whitworth threads under
static tension,” Engineering, 168, 1949 July 22nd, 93–95: Machinery, 74, 1949
June 16th, 817: Machinist, 1949 June 18th, 267.
(17) Gray, R. M., Preliminary investigation of load distribution in threaded connections,
Report WAL 730/562-47 of Division of Engineering Sciences Laboratory, Purdue Research Foundation,
Indiana, U.S.A., 1954.
(18) Cullimore, M. S. G., “Basic factors in the behaviour of friction-grip bolt joints, Part 1,” Civil
Engineering, 58, (680), 1963, March, 379–384.
(19) “Load-indicator washer for friction grip bolts.” Civil Engineering, 58 (680), 1963, March, 384.
(20) Ball, E. F. & Higgins, J. J., “Installation and tightening of bolts,” Trans, Am. Soc. Civil
Engineers, 126 (2), 1961,797–810.
(21) “The endurance of bolts. (Steel material — tensile loading)”, Roy. Aero. Soc. Data Sheets E.06.01, 1961
October.

© BSI 12-1999 19
BS 3580:1964

BSI — British Standards Institution

BSI is the independent national body responsible for preparing
British Standards. It presents the UK view on standards in Europe and at the
international level. It is incorporated by Royal Charter.

Revisions

British Standards are updated by amendment or revision. Users of

British Standards should make sure that they possess the latest amendments or
editions.

It is the constant aim of BSI to improve the quality of our products and services.
We would be grateful if anyone finding an inaccuracy or ambiguity while using
this British Standard would inform the Secretary of the technical committee
responsible, the identity of which can be found on the inside front cover.
Tel: 020 8996 9000. Fax: 020 8996 7400.

BSI offers members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards.

Buying standards

Orders for all BSI, international and foreign standards publications should be
addressed to Customer Services. Tel: 020 8996 9001. Fax: 020 8996 7001.

In response to orders for international standards, it is BSI policy to supply the

BSI implementation of those that have been published as British Standards,
unless otherwise requested.

Information on standards

BSI provides a wide range of information on national, European and

international standards through its Library and its Technical Help to Exporters
Service. Various BSI electronic information services are also available which give
details on all its products and services. Contact the Information Centre.
Tel: 020 8996 7111. Fax: 020 8996 7048.

Subscribing members of BSI are kept up to date with standards developments

and receive substantial discounts on the purchase price of standards. For details
of these and other benefits contact Membership Administration.
Tel: 020 8996 7002. Fax: 020 8996 7001.

Copyright

Copyright subsists in all BSI publications. BSI also holds the copyright, in the
UK, of the publications of the international standardization bodies. Except as
permitted under the Copyright, Designs and Patents Act 1988 no extract may be
reproduced, stored in a retrieval system or transmitted in any form or by any
means – electronic, photocopying, recording or otherwise – without prior written
permission from BSI.

This does not preclude the free use, in the course of implementing the standard,
of necessary details such as symbols, and size, type or grade designations. If these
details are to be used for any other purpose than implementation then the prior
written permission of BSI must be obtained.

If permission is granted, the terms may include royalty payments or a licensing

agreement. Details and advice can be obtained from the Copyright Manager.
BSI Tel: 020 8996 7070.
389 Chiswick High Road
London
W4 4AL